Virtualized computing infrastructures, commonly referred to as cloud computing sites, utilize physical servers with virtualization capabilities for hosting multiple virtual servers. A virtual server is a computer system running on emulated hardware provided by the host. An event is generally defined by a collection of virtual servers and a projected time duration and time interval during which the virtual servers are to be hosted. Events are hosted by placing their virtual servers on hosts capable of running them.
Efficient scheduling of such events is important for the host that offers hosting services for a limited time. The core utilization of the host is defined by the ability to host more events in a given time frame while maximizing profitability and fault tolerance. Some hosts, such as International Business Machine (IBM®) Business Continuity and Resiliency Services (BCRS) centers, offer a recovery service in case a disaster strikes. The recovery service allows a customer to periodically rehearse pre-implemented disaster recovery plans.
The above noted rehearsal events generally encompass the majority of event activities in said hosts and tend to consume significant administrative and infrastructure resources. It is desirable to optimize the scheduling and management of these rehearsal events. Further, in case of a recovery from a true disaster, it is desirable for the host to allocate as many resources as possible to accommodate a projected recovery plan. Said allocation of resources is often accomplished at the expense of delaying planned or on-going rehearsal events. It is desirable to minimize the disturbance to the prescheduled rehearsals while maintaining profitability.